Natural Language Processing

Researchers from Singapore developed a Signed Belief Graph Neural Network (SiBeGNN) for topic modeling in conspiratorial discourse, achieving a cDBI of 8.38 versus 13.60–67.27 for standard methods. The team analyzed 553,…

A study by researchers at the University of Vermont found that large language models (LLMs) exhibit systematic gender bias when detecting intimate partner violence (IPV), with abuse and intent detection decreasing in mix…

Researchers at the Seventh Workshop on Natural Language Processing and Computational Social Science found that emojis, while rarely offensive alone, can acquire harmful meanings through symbolic associations, sarcasm, an…

Even frontier large language models such as GPT-5.2 and leading open-weight alternatives consistently underperform fine-tuned BERT on a challenging legal reasoning task, according to a study by Caroline Cheng, Edward Sti…

A study by Maximilian Weber and Johannes B. Gruber analyzing Twitter data found that Meta's Galactica, marketed as a scientific assistant, faced rapidly escalating negative sentiment after its November 2022 release, whil…

Researchers at the University of Maryland found that the community size of early adopters and emoji semantics are positively correlated with the popularity of newly released emojis on Twitter. Analyzing a large dataset o…

Researchers Duy Dang Phu and Thìn Đặng Văn have developed an NLP framework combining relevance filtering, topic modeling, and LLM-assisted interpretation to analyze corporate strategic behavior in the Opioid Industry Doc…

Zack Proser, a WorkOS engineer who is partly hard of hearing, uses Granola in every work meeting to capture live transcripts and AI-generated notes, reducing the cognitive load of capturing meetings to a single button. T…

A developer built a benchmark to test whether a local-first knowledge tool called IWE could serve as agent memory by storing memories as linked markdown files and retrieving them with graph-aware tools. The benchmark rev…

A developer introduced Logical Word Count, a cheap, tokenizer-free measure of text length that stays roughly proportional to LLM token counts across languages, code, and mixed content. The method uses two simple rules: w…

A new library of over 90 AI prompts for email marketing provides copywriters with structured frameworks for generating subject lines, campaign emails, and personalization, organized by campaign type and job-to-be-done. T…

A developer explains how metadata filtering and reranking improve retrieval in RAG systems. Metadata filtering narrows the search space by using chunk attributes like chapter names, while cross-encoder models rerank resu…

Developer Martin Piaggi released Summarize, an open-source tool that converts videos from YouTube, TikTok, Instagram, and other sources into structured notes using AI transcription and LLM summarization. The local-first …

A developer argues that fixed-size chunking in RAG pipelines destroys document structure, which carries meaning. Instead of asking where to split a document, the developer suggests asking what the smallest unit is that s…

AI detection tools used by universities to identify AI-generated writing are unreliable and can wrongly accuse students of misconduct, according to an analysis published in Nature. Studies cited in the report found that …

At HackHazards'26, the team Git Push Pray built Dial2AI, a multilingual voice-based AI assistant that makes generative AI accessible through a regular phone call without requiring internet, a smartphone, or an app. The s…

Better Voice, a macOS app, launches as a free, on-device dictation and meeting transcription tool that sends speaker-labeled summaries to Apple Notes. The app uses local AI to clean up dictation and transcribe meetings w…

A developer built pdfmuse, a PDF/DOCX parser designed for precise text extraction in RAG pipelines. After running it on 80 real resumes, 8 failed silently with zero characters due to unhandled form XObjects in Canva expo…

Researchers at an undisclosed institution introduced MARGO, a reinforcement learning framework that mitigates thinking-induced hallucination in large reasoning models by using non-thinking rollouts as references. The met…

Researchers used large language models to extract information on bank runs from millions of digitized historical newspaper pages, creating the most comprehensive database of U.S. bank runs. The dataset, documented on a c…